Conversational text Analysis using various NLP techniques

Overview

PyConverse


Let me try first

Installation

pip install pyconverse

Usage

Please try this notebook that demos the core functionalities: basic usage notebook

Introduction

Conversation analytics plays an increasingly important role in shaping great customer experiences across various industries like finance/contact centres etc... primarily to gain a deeper understanding of the customers and to better serve their needs. This library, PyConverse is an attempt to provide tools & methods which can be used to gain an understanding of the conversations from multiple perspectives using various NLP techniques.

Why PyConverse?

I have been doing what can be called conversational text NLP with primarily contact centre data from various domains like Financial services, Banking, Insurance etc for the past year or so, and I have not come across any interesting open-source tools that can help in understanding conversational texts as such I decided to create this library that can provide various tools and methods to analyse calls and help answer important questions/compute important metrics that usually people want to find from conversations, in contact centre data analysis settings.

Where can I use PyConverse?

The primary use case is geared towards contact centre call analytics, but most of the tools that Converse provides can be used elsewhere as well.

There’s a lot of insights hidden in every single call that happens, Converse enables you to extract those insights and compute various kinds of KPIs from the point of Operational Efficiency, Agent Effectiveness & monitoring Customer Experience etc.

If you are looking to answer questions like these:-

  1. What was the overall sentiment of the conversation that was exhibited by the speakers?
  2. Was there periods of dead air(silence periods) between the agents and customer? if so how much?
  3. Was the agent empathetic towards the customer?
  4. What was the average agent response time/average hold time?
  5. What was being said on calls?

and more... pyconverse might be of small help.

What can PyConverse do?

At the moment pyconverse can do a few things that broadly fall into these categories:-

  1. Emotion identification
  2. Empathetic statement identification
  3. Call Segmentation
  4. Topic identification from call segments
  5. Compute various types of Speaker attributes:
    1. linguistic attributes like: word counts/number of words per utterance/negations etc.
    2. Identify periods of silence & interruptions.
    3. Question identification
    4. Backchannel identification
  6. Assess the overall nature of the speaker via linguistic attributes and tell if the Speaker is:
    1. Talkative, verbally fluent
    2. Informal/Personal/social
    3. Goal-oriented or Forward/future-looking/focused on past
    4. Identify inhibitions

What Next?

  1. Improve documentation.
  2. Add more use case notebooks/examples.
  3. Improve some of the functionalities and make it more streamlined.

Built with:

Transformers Spacy Pytorch

Credits:

Note: The backchannel Utterance classification method is inspired by facebook's Unsupervised Topic Segmentation of Meetings with BERT Embeddings paper (arXiv:2106.12978 [cs.LG])

You might also like...
nlabel is a library for generating, storing and retrieving tagging information and embedding vectors from various nlp libraries through a unified interface.
nlabel is a library for generating, storing and retrieving tagging information and embedding vectors from various nlp libraries through a unified interface.

nlabel is a library for generating, storing and retrieving tagging information and embedding vectors from various nlp libraries through a unified interface.

An easy-to-use framework for BERT models, with trainers, various NLP tasks and detailed annonations

FantasyBert English | 中文 Introduction An easy-to-use framework for BERT models, with trainers, various NLP tasks and detailed annonations. You can imp

Grading tools for Advanced NLP (11-711)Grading tools for Advanced NLP (11-711)

Grading tools for Advanced NLP (11-711) Installation You'll need docker and unzip to use this repo. For docker, visit the official guide to get starte

Kashgari is a production-level NLP Transfer learning framework built on top of tf.keras for text-labeling and text-classification, includes Word2Vec, BERT, and GPT2 Language Embedding.

Kashgari Overview | Performance | Installation | Documentation | Contributing 🎉 🎉 🎉 We released the 2.0.0 version with TF2 Support. 🎉 🎉 🎉 If you

Kashgari is a production-level NLP Transfer learning framework built on top of tf.keras for text-labeling and text-classification, includes Word2Vec, BERT, and GPT2 Language Embedding.

Kashgari Overview | Performance | Installation | Documentation | Contributing 🎉 🎉 🎉 We released the 2.0.0 version with TF2 Support. 🎉 🎉 🎉 If you

Using Bert as the backbone model for lime, designed for NLP task explanation (sentence pair text classification task)

Lime Comparing deep contextualized model for sentences highlighting task. In addition, take the classic explanation model "LIME" with bert-base model

Various capabilities for static malware analysis.

Malchive The malchive serves as a compendium for a variety of capabilities mainly pertaining to malware analysis, such as scripts supporting day to da

Python bindings to the dutch NLP tool Frog (pos tagger, lemmatiser, NER tagger, morphological analysis, shallow parser, dependency parser)

Frog for Python This is a Python binding to the Natural Language Processing suite Frog. Frog is intended for Dutch and performs part-of-speech tagging

pysentimiento: A Python toolkit for Sentiment Analysis and Social NLP tasks

A Python multilingual toolkit for Sentiment Analysis and Social NLP tasks

Comments
  • SemanticTextSegmentation NaN With All Stop Words

    SemanticTextSegmentation NaN With All Stop Words

    When running semantic text segmentation, I found that if the input utterance line is all stop words, (i.e. "Bye. Uh huh. Yeah."), SemanticTextSegmentation._get_similarity fails with ValueError: Input contains NaN.

    I found that adding a check for nan in both embeddings could solve this problem.

    def _get_similarity(self, text1, text2):
        sentence_1 = [i.text.strip()
                      for i in nlp(text1).sents if len(i.text.split(' ')) > 1]
        sentence_2 = [i.text.strip()
                      for i in nlp(text2).sents if len(i.text.split(' ')) > 2]
        embeding_1 = model.encode(sentence_1)
        embeding_2 = model.encode(sentence_2)
        embeding_1 = np.mean(embeding_1, axis=0).reshape(1, -1)
        embeding_2 = np.mean(embeding_2, axis=0).reshape(1, -1)
    
        if np.any(np.isnan(embeding_1)) or np.any(np.isnan(embeding_2)):
                return 1
    
        sim = cosine_similarity(embeding_1, embeding_2)
        return sim
    

    I would like to have someone else look at it because I don't want to make any assumptions that the stop words should be part of the same segments.

    opened by Haowjy 1
  • Updated  lru_cache decorator.

    Updated lru_cache decorator.

    After installing and running the library pyconverse on python-3.7 or below and using the import statement it gives error in import itself. I went through the utils file and saw that the "@lru_cache" decorator was written as per the new python(i.e. 3.8+) style hence when calling in older versions(py 3.7 and below it raises a NoneType Error) as the LRU_CACHE decorator is written as -" @lru_cache() " with paranthesis for older versions . Hence made the changes. The changes made do not cause any error on the newer versions.

    opened by AkashKhamkar 0
  • Error in importing Callyzer, SpeakerStats

    Error in importing Callyzer, SpeakerStats

    When I want to load the model it's showing this error.Whether it is currently in devloped mode des

    KeyError: "[E002] Can't find factory for 'tok2vec'. This usually happens when spaCy callsnlp.create_pipewith a component name that's not built in - for example, when constructing the pipeline from a model's meta.json. If you're using a custom component, you can write to Language.factories['tok2vec'] or remove it from the ### model meta and add it vianlp.add_pipeinstead.

    opened by kalpa277 0
Releases(v0.2.0)
  • v0.2.0(Nov 21, 2021)

    First Release of PyConverse library.

    Conversational Transcript Analysis using various NLP techniques.

    1. Emotion identification
    2. Empathetic statement identification
    3. Call Segmentation
    4. Topic identification from call segments
    5. Compute various types of Speaker attributes:
      • linguistic attributes like : word counts/number of words per utterance/negations etc
      • Identify periods of silence & interruptions.
      • Question identification
      • Backchannel identification
    6. Assess the overall nature of the speaker via linguistic attributes and tell if the Speaker is:
      • Talkative, verbally fluent
      • Informal/Personal/social
      • Goal-oriented or Forward/future-looking/focused on past
      • Identify inhibitions
    Source code(tar.gz)
    Source code(zip)
Owner
Rita Anjana
ML engineer
Rita Anjana
Proquabet - Convert your prose into proquints and then you essentially have Vogon poetry

Proquabet Turn your prose into a constant stream of encrypted and meaningless-so

Milo Fultz 2 Oct 10, 2022
构建一个多源(公众号、RSS)、干净、个性化的阅读环境

2C 构建一个多源(公众号、RSS)、干净、个性化的阅读环境 作为一名微信公众号的重度用户,公众号一直被我设为汲取知识的地方。随着使用程度的增加,相信大家或多或少会有一个比较头疼的问题——广告问题。 假设你关注的公众号有十来个,若一个公众号两周接一次广告,理论上你会面临二十多次广告,实际上会更多,运

howie.hu 678 Dec 28, 2022
[EMNLP 2021] LM-Critic: Language Models for Unsupervised Grammatical Error Correction

LM-Critic: Language Models for Unsupervised Grammatical Error Correction This repo provides the source code & data of our paper: LM-Critic: Language M

Michihiro Yasunaga 98 Nov 24, 2022
A simple Streamlit App to classify swahili news into different categories.

Swahili News Classifier Streamlit App A simple app to classify swahili news into different categories. Installation Install all streamlit requirements

Davis David 4 May 01, 2022
Convolutional 2D Knowledge Graph Embeddings resources

ConvE Convolutional 2D Knowledge Graph Embeddings resources. Paper: Convolutional 2D Knowledge Graph Embeddings Used in the paper, but do not use thes

Tim Dettmers 586 Dec 24, 2022
Yet another Python binding for fastText

pyfasttext Warning! pyfasttext is no longer maintained: use the official Python binding from the fastText repository: https://github.com/facebookresea

Vincent Rasneur 230 Nov 16, 2022
Predict an emoji that is associated with a text

Sentiment Analysis Sentiment analysis in computational linguistics is a general term for techniques that quantify sentiment or mood in a text. Can you

Tetsumichi(Telly) Umada 30 Sep 07, 2022
DomainWordsDict, Chinese words dict that contains more than 68 domains, which can be used as text classification、knowledge enhance task

DomainWordsDict, Chinese words dict that contains more than 68 domains, which can be used as text classification、knowledge enhance task。涵盖68个领域、共计916万词的专业词典知识库,可用于文本分类、知识增强、领域词汇库扩充等自然语言处理应用。

liuhuanyong 357 Dec 24, 2022
Deep Learning for Natural Language Processing - Lectures 2021

This repository contains slides for the course "20-00-0947: Deep Learning for Natural Language Processing" (Technical University of Darmstadt, Summer term 2021).

0 Feb 21, 2022
2021语言与智能技术竞赛:机器阅读理解任务

LICS2021 MRC 1. 项目&任务介绍 本项目基于官方给定的baseline(DuReader-Checklist-BASELINE)进行二次改造,对整个代码框架做了简单的重构,对核心网络结构添加了注释,解耦了数据读取的模块,并添加了阈值确认的功能,一些小的细节也做了改进。 本次任务为202

roar 29 Dec 05, 2022
A combination of autoregressors and autoencoders using XLNet for sentiment analysis

A combination of autoregressors and autoencoders using XLNet for sentiment analysis Abstract In this paper sentiment analysis has been performed in or

James Zaridis 2 Nov 20, 2021
UniSpeech - Large Scale Self-Supervised Learning for Speech

UniSpeech The family of UniSpeech: WavLM (arXiv): WavLM: Large-Scale Self-Supervised Pre-training for Full Stack Speech Processing UniSpeech (ICML 202

Microsoft 281 Dec 15, 2022
DeLighT: Very Deep and Light-Weight Transformers

DeLighT: Very Deep and Light-weight Transformers This repository contains the source code of our work on building efficient sequence models: DeFINE (I

Sachin Mehta 440 Dec 18, 2022
Course project of [email protected]

NaiveMT Prepare Clone this repository git clone [email protected]:Poeroz/NaiveMT.git

Poeroz 2 Apr 24, 2022
Training and evaluation codes for the BertGen paper (ACL-IJCNLP 2021)

BERTGEN This repository is the implementation of the paper "BERTGEN: Multi-task Generation through BERT" (https://arxiv.org/abs/2106.03484). The codeb

<a href=[email protected]"> 9 Oct 26, 2022
:mag: Transformers at scale for question answering & neural search. Using NLP via a modular Retriever-Reader-Pipeline. Supporting DPR, Elasticsearch, HuggingFace's Modelhub...

Haystack is an end-to-end framework that enables you to build powerful and production-ready pipelines for different search use cases. Whether you want

deepset 6.4k Jan 09, 2023
Neural network models for joint POS tagging and dependency parsing (CoNLL 2017-2018)

Neural Network Models for Joint POS Tagging and Dependency Parsing Implementations of joint models for POS tagging and dependency parsing, as describe

Dat Quoc Nguyen 152 Sep 02, 2022
Implementation of Natural Language Code Search in the project CodeBERT: A Pre-Trained Model for Programming and Natural Languages.

CodeBERT-Implementation In this repo we have replicated the paper CodeBERT: A Pre-Trained Model for Programming and Natural Languages. We are interest

Tanuj Sur 4 Jul 01, 2022
Implementation of Multistream Transformers in Pytorch

Multistream Transformers Implementation of Multistream Transformers in Pytorch. This repository deviates slightly from the paper, where instead of usi

Phil Wang 47 Jul 26, 2022
BERT-based Financial Question Answering System

BERT-based Financial Question Answering System In this example, we use Jina, PyTorch, and Hugging Face transformers to build a production-ready BERT-b

Bithiah Yuan 61 Sep 18, 2022